itemtype for mixedmirt

[Keith Lau]

Hi Phil,

Does mixedmirt support created itemtype? Manual seems not mention that. Thanks.

[Phil Chalmers]

Hey Keith,

No it does not, mostly because it would be too hard to track intercepts. It's possible that latent regression models could be supported in the future though (this might work in mirt(), but I'm not 100% sure), but currently they are not supported in mixedmirt(). Cheers.

Phil

On Fri, Jun 24, 2016 at 7:56 AM, Keith Lau <genw...@gmail.com> wrote:

Hi Phil,

Does mixedmirt support created itemtype? Manual seems not mention that. Thanks.

--
You received this message because you are subscribed to the Google Groups "mirt-package" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mirt-package...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Keith Lau]

Hi Phil,

How about letting default itemtype being modifiable (e.g.,  adding new parameter to "gpcm" or "nominal") with the default intercepts of items "d" preserved. I would like to do so to estimate the rater effect. So, the mixedmirt() can be used. Thanks!




Phil Chalmers於 2016年6月24日星期五 UTC+8下午11時28分38秒寫道：

[Phil Chalmers]

Hi Keith,

The rater effect idea for polytomous models is certainly something I'm considering for the future in mixedmirt, but I'm a bit too busy these days to really experiment with it. So it's on the todo list, but I wouldn't expect it any time soon.

Phil

[Keith Lau]

Hi, Phil. I have experimented rater effect just creating a new parameter "rater" for each item, estimated by mirt(). It is quite straightforward, so mixedmirt() should be capable as well. And thank you for this efforts.
 
Phil Chalmers於 2016年6月26日星期日 UTC+8上午4時38分31秒寫道：